Omega-3 long-chain polyunsaturated fatty acid(s) (LC-PUFA) are now widely recognized as important compounds for human and animal health. These fatty acids may be obtained from dietary sources or by conversion of linoleic (LA, omega-6) or α-linolenic (ALA, omega-3) fatty acids, both of which are regarded as essential fatty acids in the human diet. While humans and many other vertebrate animals are able to convert LA or ALA, obtained from plant sources, to LC-PUFA, they carry out this conversion at a very low rate. Moreover, most modern societies have imbalanced diets in which at least 90% of polyunsaturated fatty acid(s) (PUFA) consist of omega-6 fatty acids, instead of the 4:1 ratio or less for omega-6:omega-3 fatty acids that is regarded as ideal (Trautwein, 2001). The immediate dietary source of LC-PUFA such as eicosapentaenoic acid (EPA, 20:5) and docosahexaenoic acid (DHA, 22:6) for humans is mostly from fish or fish oil. Health professionals have therefore recommended the regular inclusion of fish containing significant levels of LC-PUFA into the human diet. Increasingly, fish-derived LC-PUFA oils are being incorporated into food products and in infant formula. However, due to a decline in global and national fisheries, alternative sources of these beneficial health-enhancing oils are needed.
Inclusion of omega-3 LC-PUFA such as EPA and DHA in the human diet has been linked with numerous health-related benefits. These include prevention or reduction of coronary heart disease, hypertension, type-2 diabetes, renal disease, rheumatoid arthritis, ulcerative colitis and chronic obstructive pulmonary disease, and aiding brain development and growth (Simopoulos, 2000). More recently, a number of studies have also indicated that omega-3 PUFA may be beneficial in infant nutrition and development and against various mental disorders such as schizophrenia, attention deficit hyperactive disorder and Alzheimer's disease.
Higher plants, in contrast to animals, lack the capacity to synthesise polyunsaturated fatty acids with chain lengths longer than 18 carbons. In particular, crop and horticultural plants along with other angiosperms do not have the enzymes needed to synthesize the longer chain omega-3 fatty acids such as EPA, DPA and DHA that are derived from ALA. An important goal in plant biotechnology is therefore the engineering of crop plants, particularly oilseed crops, that produce substantial quantities of LC-PUFA, thus providing an alternative source of these compounds.
Pathways of LC-PUFA Synthesis
Biosynthesis of LC-PUFA from linoleic and α-linolenic fatty acids in organisms such as microalgae, mosses and fungi may occur by a series of alternating oxygen-dependent desaturations and elongation reactions as shown schematically in FIG. 1. In one pathway (FIG. 1, II), the desaturation reactions are catalysed by Δ6, Δ5, and Δ4 desaturases, each of which adds an additional double bond into the fatty acid carbon chain, while each of a Δ6 and a Δ5 elongase reaction adds a two-carbon unit to lengthen the chain. The conversion of ALA to DHA in these organisms therefore requires three desaturations and two elongations. Genes encoding the enzymes required for the production of DHA in this aerobic pathway have been cloned from various microorganisms and lower plants including microalgae, mosses, fungi. Genes encoding some of the enzymes including one that catalyses the fifth step, the Δ5 elongase, have been isolated from vertebrate animals including mammals (reviewed in Sayanova and Napier, 2004). However, the Δ5 elongase isolated from human cells is not specific for the EPA to DPA reaction, having a wide specificity for fatty acid substrates (Leonard et al., 2002).
Alternative routes have been shown to exist for two sections of the ALA to DHA pathway in some groups of organisms. The conversion of ALA to ETA may be carried out by a combination of a Δ9 elongase and a Δ8 desaturase (the so-called Δ8 desaturation route, see FIG. 1, IV) in certain protists and thraustochytrids, as evidenced by the isolated of genes encoding such enzymes (Wallis and Browse, 1999; Qi et al., 2002). In mammals, the so-called “Sprecher” pathway converts DPA to DHA by three reactions, independent of a Δ4 desaturase (Sprecher et al., 1995).
Besides these desaturase/elongase systems, EPA and DHA can also be synthesized through an anaerobic pathway in a number of organisms such as Shewanella, Mortiella and Schizhochytrium (Abbadi et al., 2001). The operons encoding these polyketide synthase (PKS) enzyme complexes have been cloned from some bacteria (Morita et al., 2000; Metz et al., 2001; Tanaka et al., 1999; Yazawa, 1996; Yu et al., 2000; WO 00/42195). The EPA PKS operon isolated from Shewanella spp has been expressed in Synechococcus allowing it to synthesize EPA (Takeyama et al., 1997). The genes encoding these enzymes are arranged in relatively large operons, and their expression in transgenic plants has not been reported. Therefore it remains to be seen if the anaerobic PKS-like system is a possible alternative to the more classic aerobic desaturase/elongase for the transgenic synthesis of LC-PUFA.
Desaturases
The desaturase enzymes that have been shown to participate in LC-PUFA biosynthesis all belong to the group of so-called “front-end” desaturases which are characterised by the presence of a cytochrome b5 domain at the N-terminus of each protein. The cyt b5 domain presumably acts as a receptor of electrons required for desaturation (Napier et al., 1999; Sperling and Heinz, 2001).
The enzyme Δ5 desaturase catalyses the further desaturation of C20 LC-PUFA leading to arachidonic acid (ARA, 20:4ω6) and EPA (20:5ω3). Genes encoding this enzyme have been isolated from a number of organisms, including algae (Thraustochytrium sp. Qiu et al., 2001), fungi (M. alpine, Pythium irregulare, Michaelson et al., 1998; Hong et al., 2002), Caenorhabditis elegans and mammals. A gene encoding a bifunctional Δ5−/Δ6−desaturase has also been identified from zebrafish (Hasting et al., 2001). The gene encoding this enzyme might represent an ancestral form of the “front-end desaturase” which later duplicated and evolved distinct functions. The last desaturation step to produce DHA is catalysed by a Δ4 desaturase and a gene encoding this enzyme has been isolated from the freshwater protist species Euglena gracilis and the marine species Thraustochytrium sp. (Qiu et al., 2001; Meyer et al., 2003).
Elongases
Several genes encoding PUFA-elongation enzymes have also been isolated (Sayanova and Napier, 2004). The members of this gene family were unrelated to the elongase genes present in higher plants, such as FAE1 of Arabidopsis, that are involved in the extension of saturated and monounsaturated fatty acids. An example of the latter is erucic acid (22:1) in Brassicas. In some protist species, LC-PUFA are synthesized by elongation of linoleic or α-linolenic acid with a C2 unit, before desaturation with Δ8 desaturase (FIG. 1 part IV; “Δ8-desaturation” pathway). Δ6 desaturase and Δ6, elongase activities were not detected in these species. Instead, a Δ9-elongase activity would be expected in such organisms, and in support of this, a C18 Δ9-elongase gene has recently been isolated from Isochrysis galbana (Qi et al., 2002).
Engineered Production of LC-PUFA
Transgenic oilseed crops that are engineered to produce major LC-PUFA by the insertion of these genes have been suggested as a sustainable source of nutritionally important fatty acids. However, the requirement for coordinate expression and activity of five new enzymes encoded by genes from possibly diverse sources has made this goal difficult to achieve and the proposal remained speculative until now.
The LC-PUFA oxygen-dependent biosynthetic pathway to form EPA (FIG. 1) has been successfully constituted in yeast by the co-expression of a Δ6-elongase with Δ6- and Δ5 fatty acid desaturases, resulting in small but significant accumulation of ARA and EPA from exogenously supplied linoleic and α-linolenic acids (Beaudoin et al., 2000; Zank et al., 2000). This demonstrated the ability of the genes belonging to the LC-PUFA synthesis pathway to function in heterologous organisms. However, the efficiency of producing EPA was very low. For example, three genes obtained from C. elegans, Borago officinalis and Mortierella alpina were expressed in yeast (Beaudoin et al., 2000). When the transformed yeast were supplied with 18:2ω-3 (LA) or 18:3ω-3 (ALA), there was slight production of 20:4ω-6 or 20:5ω-3, at conversion efficiencies of 0.65% and 0.3%, respectively. Other workers similarly obtained very low efficiency production of EPA by using genes expressing two desaturases and one elongase in yeast (Domergue et al., 2003a; Zank et al., 2002). There remains, therefore, a need to improve the efficiency of production of EPA in organisms such as yeast, let alone the production of the C22 PUFA which requires the provision of additional enzymatic steps.
Some progress has been made in the quest for introducing the aerobic LC-PUFA biosynthetic pathway into higher plants including oilseed crops (reviewed by Sayanova and Napier, 2004; Drexler et al., 2003; Abbadi et al., 2001). A gene encoding a Δ6-fatty acid desaturase isolated from borage (Borago officinalis) was expressed in transgenic tobacco and Arabidopsis, resulting in the production of GLA (18:3ω6) and SDA (18:4ω3), the direct precursors for LC-PUFA, in the transgenic plants (Sayanova et al., 1997; 1999). However, this provides only a single, first step.
Domergue et al. (2003a) used a combination of three genes, encoding Δ6- and Δ5 fatty acid desaturases and a Δ6-elongase in both yeast and transgenic linseed. The desaturase genes were obtained from the diatom Phaeodactylum tricornutum and the elongase gene from the moss Physcomitrella patens. Low elongation yields were obtained for endogenously produced Δ6-fatty acids in yeast cells (i.e. combining the first and second enzymatic steps), and the main C20 PUFA product formed was 20:2Δ11,14, representing an unwanted side reaction. Domergue et al. (2003a) also state, without presenting data, that the combination of the three genes were expressed in transgenic linseed which consequently produced ARA and EPA, but that production was inefficient. They commented that the same problem as had been observed in yeast existed in the seeds of higher plants and that the “bottleneck” needed to be circumvented for production of LC-PUFA in oil seed crops.
WO 2004/071467 (DuPont) reported the expression of various desaturases and elongases in soybean cells but did not show the synthesis of DHA in regenerated plants or in seeds.
Abbadi et al. (2004) described attempts to express combinations of desaturases and elongases in transgenic linseed, but achieved only low levels of synthesis of EPA. Abbadi et al. (2004) indicated that their low levels of EPA production were also due to an unknown “bottleneck”.
Qi et al. (2004) achieved synthesis in leaves but did not report results in seeds. This is an important issue as the nature of LC-PUFA synthesis can vary between leaves and seeds. In particular, oilseeds store lipid in seeds mostly as TAG while leaves synthesize the lipid mostly as phosphatidyl lipids. Furthermore, Qi et al. (2004) only produced AA and EPA.
As a result, there is a need for further methods of producing long-chain polyunsaturated, particularly EPA, DPA and DHA, in recombinant cells.